CN116824837A - Data acquisition method, dynamic interest point identification method and device - Google Patents

Abstract

The application provides a data acquisition method, a dynamic interest point identification method and a device, wherein the dynamic interest point identification method comprises the following steps: receiving first steering behavior related data sent by a first vehicle during occurrence of steering behavior; inputting the first steering behavior related data into a first model, and obtaining a lane change behavior type corresponding to the first steering behavior related data, wherein the lane change behavior type comprises a first lane change type used for improving the speed of a vehicle and a second lane change type used for triggering lane change behavior due to the existence of a dynamic interest point; under the condition that the corresponding lane change behavior type of the first steering behavior related data is the second lane change type, the first steering behavior related data is input into the second model, and the type and the position coordinates of the first dynamic interest point output by the second model are obtained. The method can more accurately identify various dynamic interest points, assist the vehicle to make pre-judgment and planning adjustment in advance, reduce the calculated amount of the vehicle end and promote the driving safety and riding comfort experience.

Description

Data acquisition method, dynamic interest point identification method and device

Technical Field

The application relates to the technical field of geographic information, in particular to a data acquisition method, a dynamic interest point identification method and a dynamic interest point identification device.

Background

Points of interest (Point of Interest, POI) is a term in geographic information systems that generally refers to everything that can be abstracted into points, especially some geographic entities that are closely related to people's life, such as schools, banks, restaurants, gas stations, hospitals, supermarkets, and the like. The main purpose of the interest points is to describe the addresses of things or events, so that the description capability and the inquiry capability of the positions of the things or events can be enhanced to a great extent, and the accuracy and the speed of geographic positioning are improved.

The interest point of the geographic space, such as the ramp merging position of the highway construction area shown in fig. 1 and 2, is usually set as a "construction area merging section" < - "interest point" in the highway construction. When a large vehicle (such as vehicle C) is present in the merging section, in fig. 1, the vehicle C and the vehicle B run close to each other in adjacent lanes, and the head of the vehicle C is in front of the vehicle B. Since the lane in which the vehicle C is located is in road construction, the vehicle C may be suddenly changed to the lane in which the vehicle B is located due to restriction reasons such as a braking reaction time of a large vehicle and a blind observation area, and thus a lateral collision is easily caused (as shown in fig. 2).

The probability and the severity of traffic collision of the ramp confluence road section are higher than those of the transfer road section, so that measures can be taken pertinently to reduce the traffic collision probability so as to improve the traffic safety level, such as shunting of a large vehicle; taking control measures on a specific road section to prohibit overtaking, setting a vehicle distance confirmation facility and the like.

The POI data acquisition has the following four characteristics:

1. the threshold is high: only specific mapping organizations can collect geographic information;

2. the cost is high: the number of POIs is large and widely distributed, and a large amount of manpower and financial resources are required to be consumed for acquisition;

3. time consuming: POI data collection takes a lot of time to collect;

4. difficult maintenance: the real world changes rapidly, the POI data acquisition period is long, and the POI data acquisition period is difficult to ensure to be consistent with the real world.

The following schemes are generally used for mining points of interest in the prior art:

1. and shooting an image, and detecting whether the interest points are matched or not by comparing semantic analysis with an interest point/non-interest point sensitive word stock by adopting a detection module based on natural language processing (Natual Language Processing, NLP) for the shot image. The image detection model may also be a neural network model. The proposal extracts the acquired text information in the image to be detected; based on the text information in the image to be detected, a POI detection module based on NLP is adopted to detect whether the image to be detected is a POI image.

2. Dividing a geographic space in a grid form, compressing check-in data of a user in a range of n square meters into a grid point, wherein the check-in quantity refers to the number of times that an area contained in the grid is checked in by the user; searching any grid, determining probability containing POI, and performing correlation logic operation to obtain final probability. The scheme determines each grid contained in a target area corresponding to the target service; determining the inclusion probability of any one grid including POI (point of interest) aiming at any one grid; taking grids with probability larger than a first preset threshold as candidate POI grids; and determining the POIs in the target area according to the POIs contained in each candidate POI grid, so as to execute the related logic of the target service according to the POIs in the target area.

The above method for mining the points of interest is based on static (unchanged in a short time) geographic information, and dynamic mining of the points of interest has not been considered.

Dynamic points of interest refer to points of interest that dynamically appear as a function of time and space, typically occurring and existing in a short period of time due to an emergency. The short time may be a point of interest that is less than a predetermined time period threshold (e.g., minutes, hours, days, etc.) relative to the time of existence of the static point of interest. For example, the junction position of the ramp in the highway construction area, the congestion point caused by the occasional traffic accident, the congestion point caused by the sudden increase of the traffic flow, the congestion of the urban expressway exit near the hot-spot destination (for example, a hospital) at a regular period, and the like. The corresponding time thresholds may be different for different types of dynamic points of interest.

A mining method of dynamic interest points utilizes a mode of combining millimeter wave radar and video to detect traffic accident events, and realizes all-weather automatic detection and early warning reminding of expressway events. Specifically, a data acquisition and analysis method is adopted, whether the vehicle has an accident or not is judged through analysis, and corresponding accident warning measures are adopted.

Another method for mining dynamic points of interest, comprising: step one, acquiring real-time GPS data information of a floating car, wherein the real-time GPS data information comprises car ID, GPS time, speed, direction and longitude and latitude coordinate information; step two, matching the GPS points of the vehicles to the road links to generate vehicle track data based on the road network; step three, establishing a high-speed and rapid route accident risk index model of single car rear-end collision of a congestion road section caused by accidental accidents; determining the position of the congestion team tail which can cause the rear-end collision accident according to the running track of the multiple vehicles; step five: the identified and consecutive links are integrated together.

The mining method of the dynamic interest points is based on analysis of image and video data, and the analysis based on GPS+speed data is mainly known. The former is large in calculation amount (high in realization difficulty) and suitable for judging the running state of a single vehicle or a few vehicles, and the latter is limited in type of interest points and can only recognize congestion.

Disclosure of Invention

The application aims to solve the technical problems of providing a data acquisition method, a dynamic interest point identification method and a dynamic interest point identification device, which can assist a vehicle to make pre-judgment and planning adjustment in advance, reduce the calculated amount of a vehicle end, or more accurately identify more kinds of dynamic interest points and improve driving safety and riding comfort experience.

In order to solve the technical problems, the embodiment of the application provides the following technical scheme:

in a first aspect, an embodiment of the present application provides a data acquisition method, including:

collecting and storing steering behavior related data of the vehicle during the occurrence of steering behavior;

after the steering behavior is finished, judging whether the steering behavior accords with a preset condition;

and under the condition that the steering behavior accords with a preset condition, transmitting steering behavior related data during the occurrence of the steering behavior to a server.

Optionally, the steering behavior related data includes at least one of the following data:

vehicle track data including track start time, end time, track length, geographic location point timing of the track;

steering behavior data including steering duration, steering angle of steering wheel, geographic location when turning on and off steering lights;

speed-related data including a vehicle speed during occurrence of steering behavior, an opening degree when an accelerator pedal opening degree is greater than a first threshold value, an opening degree when a decelerator pedal opening degree is greater than a second threshold value, a duration for which a speed is lower than a third threshold value;

time and environmental time, including date, time of day, weather information and temperature.

Optionally, the preset condition includes at least one of the following conditions: the steering angle is smaller than a preset angle value; the steering position is not a preset area.

Optionally, the method further comprises:

and deleting the steering behavior related data during the occurrence of the steering behavior under the condition that the steering behavior does not accord with a preset condition.

Optionally, the sending the steering behavior related data during the occurrence of the steering behavior to a server includes: and when the vehicle speed is greater than the average vehicle speed corresponding to the current road type of the vehicle, transmitting the vehicle identification and the steering behavior related data during the occurrence of the steering behavior to a server.

In a second aspect, an embodiment of the present application provides a method for identifying a dynamic point of interest, including:

receiving first steering behavior related data sent by a first vehicle during occurrence of steering behavior;

inputting the first steering behavior related data into a first model for identifying a lane change behavior type, and obtaining a lane change behavior type corresponding to the first steering behavior related data, wherein the lane change behavior type comprises a first lane change type for improving the vehicle speed and a second lane change type for triggering the lane change behavior due to the existence of a dynamic interest point;

and under the condition that the corresponding track behavior type of the first steering behavior related data is the second track changing type, inputting the first steering behavior related data into a second model for classifying the dynamic interest points, and obtaining the type and the position coordinates of the first dynamic interest points output by the second model.

Optionally, the method further comprises:

and sending the type and the position coordinates of the first dynamic interest point to the related vehicle. Wherein the associated vehicle comprises at least one of: the distance between the first dynamic interest point and the vehicle is smaller than a preset distance, and other member vehicles in a vehicle team where the first vehicle is located; a navigation path passes through the vehicle of the first dynamic point of interest.

Optionally, the method further comprises:

performing supervised machine learning model training by utilizing pre-collected steering behavior related data corresponding to different lane change types to obtain the first model;

and performing supervised machine learning model training by utilizing steering behavior related data which are collected in advance and correspond to different dynamic interest points, so as to obtain the second model.

In a third aspect, an embodiment of the present application provides a data acquisition device, including:

the acquisition module is used for acquiring and storing steering behavior related data of the vehicle during the occurrence of steering behavior;

the judging module is used for judging whether the steering behavior accords with a preset condition after the steering behavior is finished;

and the sending module is used for sending the steering behavior related data during the occurrence period of the steering behavior to a server under the condition that the steering behavior accords with a preset condition.

Optionally, the steering behavior related data includes at least one of the following data:

vehicle track data including track start time, end time, track length, geographic location point timing of the track;

steering behavior data including steering duration, steering angle of steering wheel, geographic location when turning on and off steering lights;

speed-related data including a vehicle speed during occurrence of steering behavior, an opening degree when an accelerator pedal opening degree is greater than a first threshold value, an opening degree when a decelerator pedal opening degree is greater than a second threshold value, a duration for which a speed is lower than a third threshold value;

time and environmental data including date, time of day, weather information and temperature.

Optionally, the preset condition includes at least one of the following conditions: the steering angle is smaller than a preset angle value; the steering position is not a preset area.

Optionally, the method further comprises:

and the deleting module is used for deleting the steering behavior related data during the occurrence period of the steering behavior under the condition that the steering behavior does not accord with the preset condition.

Optionally, the sending module is further configured to send, to the server, the vehicle identifier and the steering behavior related data during the occurrence of the steering behavior when the vehicle speed is greater than an average vehicle speed corresponding to a road type on which the vehicle is currently located.

In a fourth aspect, an embodiment of the present application provides a data acquisition device, including: a processor, a memory and a program stored on the memory and executable on the processor, which when executed by the processor implements the steps of the method as described in the first aspect.

In a fifth aspect, an embodiment of the present application provides a server including:

the receiving module is used for receiving first steering behavior related data sent by the first vehicle during the occurrence of steering behavior;

the first recognition module is used for inputting the first steering behavior related data into a first model for recognizing the lane change behavior type, and obtaining the lane change behavior type corresponding to the first steering behavior related data, wherein the lane change behavior type comprises a first lane change type for improving the vehicle speed and a second lane change type for triggering the lane change behavior due to the existence of a dynamic interest point;

the second recognition module is used for inputting the first steering behavior related data into a second model for classifying the dynamic interest points under the condition that the behavior type corresponding to the first steering behavior related data is a second lane change type, and obtaining the type and the position coordinates of the first dynamic interest points output by the second model.

Optionally, the method further comprises:

and the sending module is used for sending the type and the position coordinates of the first dynamic interest point to the related vehicle. Wherein the associated vehicle comprises at least one of: the distance between the first dynamic interest point and the vehicle is smaller than a preset distance, and other member vehicles in a vehicle team where the first vehicle is located; a navigation path passes through the vehicle of the first dynamic point of interest.

Optionally, the method further comprises:

the first training module is used for training a supervised machine learning model by utilizing steering behavior related data which are collected in advance and correspond to different lane change types to obtain the first model;

and the second training module is used for performing supervised machine learning model training by utilizing the steering behavior related data which are collected in advance and correspond to different dynamic interest points, so as to obtain the second model.

In a sixth aspect, an embodiment of the present application provides a server, including: a processor, a memory and a program stored on the memory and executable on the processor, which when executed by the processor implements the steps of the method as described in the second aspect.

In a seventh aspect, an embodiment of the present application provides a dynamic point of interest identification system, including at least one data acquisition device according to the third aspect, and a server according to the fifth aspect.

In an eighth aspect, embodiments of the present application provide a computer readable storage medium having stored thereon a program which, when executed by a processor, implements the steps of the method according to the first or second aspect.

Compared with the prior art, the data acquisition method, the dynamic interest point identification method and the dynamic interest point identification device provided by the embodiment of the application acquire steering behavior related data of the vehicle during the occurrence period of steering behaviors through the vehicle end, and the server identifies the dynamic interest points, so that the vehicle can be assisted to conduct prejudgment and planning adjustment in advance, the calculated amount of the vehicle end is reduced, more kinds of dynamic interest points can be identified more accurately, and the driving safety and riding comfort experience are improved.

Drawings

FIG. 1 is a schematic illustration of a ramp merge of a highway construction area;

FIG. 2 is another schematic illustration of a ramp merge of a highway construction zone;

FIG. 3 is a schematic flow chart of a data acquisition method according to an embodiment of the application;

FIG. 4 is a flowchart of a method for identifying dynamic points of interest according to an embodiment of the present application;

FIG. 5 is an exemplary diagram of a data acquisition method according to an embodiment of the present application;

FIG. 6 is an exemplary diagram of a method for identifying dynamic points of interest according to an embodiment of the present application;

FIG. 7 is a schematic view showing an application effect of the method according to the embodiment of the present application

FIG. 8 is a schematic diagram of a data acquisition device according to an embodiment of the present application;

FIG. 9 is a schematic diagram of another structure of a data acquisition device according to an embodiment of the present application;

FIG. 10 is a schematic diagram of a server according to an embodiment of the present application;

fig. 11 is a schematic diagram of another structure of a server according to an embodiment of the present application.

Detailed Description

In order to make the technical problems, technical solutions and advantages to be solved by the embodiments of the present application more apparent, the following detailed description will be given with reference to the accompanying drawings and the specific embodiments.

The terms first, second and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein.

The embodiment of the application provides a dynamic interest point identification system, which utilizes steering behavior related data acquired by a vehicle end during the occurrence of steering behavior of a vehicle to identify dynamic interest points at a server end, so that the calculated amount of the vehicle end can be reduced, more various dynamic interest points can be identified more accurately, and further the driving safety can be assisted to be improved.

Referring to fig. 3, an embodiment of the present application provides a data collection method, where the method applies a data collection module of a vehicle end, and specifically includes:

step 31, collecting and storing steering behavior related data of the vehicle during occurrence of steering behavior.

Here, the steering behavior-related data generally includes at least one of the following data:

(1) Vehicle track data including track start time, track end time, track length, track geographic location point timing (e.g., GPS location timing), etc.

(2) Steering behavior data including steering duration, steering angle of the steering wheel, geographic location when turning on and off the steering lamp, etc. Here, turning on and off the turn signal may be used as starting and ending time points of the turning behavior. The track start time and the track end time may be the times when the turn signal is turned on and off, respectively.

(3) The speed-related data includes a vehicle speed during occurrence of steering behavior, an opening degree when an accelerator pedal opening degree is greater than a first threshold value, an opening degree when a decelerator pedal opening degree is greater than a second threshold value, a duration for which the speed is lower than a third threshold value, and the like. Here, the first, second, and third thresholds may be set in size as needed.

(4) Time and environmental data including date, time of day, weather information, temperature, etc. Here, the date may be a specific year, month, and day. The moment of time represents a certain time or period of time within a day, such as the time when the steering action starts (which may specifically be the time when the turn light is turned on or the time when the steering angle is greater than a certain preset threshold). The weather information may be information reflecting weather conditions, and the temperature indicates air temperature.

The above data may be obtained from the relevant sensors or vehicle control system, typically by accessing the vehicle CAN bus. The data acquisition module acquires the data and temporarily stores the data locally.

In addition, it should be noted that, in order to obtain more reference data, the period from the start of the steering behavior to the end of the steering behavior of the vehicle according to the embodiment of the present application may include not only a period of time before the start of the steering behavior (such as a period of time from 20 seconds before the start of the steering behavior to the start of the steering behavior) and a period of time after the end of the steering behavior (such as a period of time from the end of the steering behavior to 20 seconds after the end of the steering behavior) during the occurrence of the steering behavior.

When the steering behavior related data is collected, the time period before the steering behavior starts and the time period after the steering behavior ends can be collected

And step 32, judging whether the steering behavior accords with a preset condition after the steering behavior is finished.

And step 33, transmitting the steering behavior related data during the occurrence of the steering behavior to a server under the condition that the steering behavior accords with a preset condition.

Here, after the steering behavior of the vehicle is completed, the data acquisition module determines whether the steering behavior meets a preset condition. Specifically, the preset conditions include at least one of the following conditions: 1) The steering angle is smaller than a preset angle value; 2) The steering position is not a preset area.

The steering angle is larger than a preset angle value (such as 90 degrees), which means the maximum steering angle in steering behavior, and can be obtained specifically by the maximum change angle of the head direction during steering behavior. When the vehicle passes through a left-hand lane, a right-hand lane or a U-bend, although the vehicle experiences steering behavior, these steering behaviors are typically static curve areas on the road with less relevance to the dynamic point of interest, and therefore, embodiments of the present application may not take such steering behavior into account. In addition, steering behavior typically occurs when the vehicle exits the highway through the highway ramp, and these regions are also static curve regions, so similar regions may be set as preset regions, and embodiments of the present application may not consider steering behavior occurring in the preset regions.

In this way, through the steps 32 to 33, the embodiment of the application can filter out the steering behavior related to the static interest point, and only send the steering behavior related data possibly related to the dynamic interest point to the server for processing, so as to avoid sending unnecessary data, and reduce the data sending amount and the local data storage amount.

In the embodiment of the present application, when the steering behavior does not meet a preset condition, for example, when the steering angle is greater than or equal to the preset angle value, or when the steering position belongs to the preset area, the locally stored steering behavior related data during the occurrence of the steering behavior is deleted.

Considering that the calculation amount of the controller of the vehicle at the dynamic interest point is generally larger, for example, in the case of automatic driving, the controller needs more calculation resources to process the traffic congestion condition of the dynamic interest point, so in order to reduce the occupation of the calculation resources in the case, the embodiment of the application can send the vehicle identifier and the steering behavior related data during the occurrence period of the steering behavior to the server when the vehicle speed is greater than the average vehicle speed corresponding to the current road type of the vehicle. The average vehicle speed may be an average vehicle speed obtained by counting the same type of road.

Through the method, the data acquisition device provided by the embodiment of the application can send the steering behavior related data possibly related to the dynamic interest point to the server so as to help the server determine whether the dynamic interest point exists or not based on the data.

Referring to fig. 4, the embodiment of the present application further provides a method for identifying a dynamic interest point, when applied to a server, the method includes:

step 41, receiving first steering behavior related data during occurrence of steering behavior transmitted by a first vehicle.

Here, the first vehicle may obtain the first steering behavior-related data during occurrence of the steering behavior and transmit the data to the server with reference to the data acquisition method described in fig. 3. The server receives the data and performs the following processing.

And 42, inputting the first steering behavior related data into a first model for identifying the lane change behavior type, and obtaining the lane change behavior type corresponding to the first steering behavior related data, wherein the lane change behavior type comprises a first lane change type for improving the vehicle speed and a second lane change type for triggering the lane change behavior due to the existence of a dynamic interest point.

Here, the input data of the first model is steering behavior-related data, and the output data is a lane change behavior type. In general, the first lane change type (ordinary lane change behavior) for increasing the vehicle speed is such that the vehicle speed is significantly increased after the steering behavior occurs, and the steering behavior-related data indicates that the accelerator opening is greater than a certain preset threshold value or the speed increase ratio is greater than a certain preset ratio after the steering behavior is ended. That is, the first lane change type is a lane change type that makes a lane change for the purpose of increasing the vehicle speed. Unlike the first lane change type, the second lane change type (temporary lane change behavior) that triggers lane change behavior due to the presence of a dynamic point of interest, the vehicle speed is typically not significantly increased after the steering behavior occurs. By the above distinction, different lane change types can be identified. The embodiment of the application can utilize the steering behavior related data which are collected in advance and correspond to different lane changing types as training data to perform supervised machine learning model training so as to obtain the first model. The training data are easy to obtain, for example, from automobile manufacturing enterprises, automatic driving technical enterprises and even open source simulation platforms.

And step 43, inputting the first steering behavior related data to a second model for classifying the dynamic interest points under the condition that the lane change behavior type corresponding to the first steering behavior related data is a second lane change type, and obtaining the type and the position coordinates of the first dynamic interest points output by the second model.

Here, the input data of the second model is steering behavior related data, and the output data is the type and position coordinates of the dynamic interest point. In the embodiment of the present application, the types of the dynamic interest points include, but are not limited to, the following types: the ramp merging position of the expressway construction area, the congestion point caused by occasional traffic accidents, the congestion point caused by sudden increase of traffic flow, the congestion of urban expressway exits near hot destinations (such as hospitals) at regular time intervals, and the like. The steering behavior related data corresponding to the different types of dynamic interest points also have respective characteristics, and the embodiment of the application can utilize the steering behavior related data which is collected in advance and corresponds to the different dynamic interest points as training data to perform supervised machine learning model training so as to obtain the second model. The training data may be implemented by an autopilot simulator.

Through the steps, the embodiment of the application can more accurately identify more kinds of dynamic interest points by utilizing the steering behavior related data collected by the vehicle end, assist the vehicle to make pre-judgment and planning adjustment in advance, and assist in improving driving safety and riding comfort experience while reducing the calculated amount of the vehicle end.

In the embodiment of the application, the supervised machine learning method used by the first model and the second model can be a random forest method and a support vector machine method, including but not limited to other machine learning methods suitable for driving behavior and interest point type judgment. According to the embodiment of the application, the interest point classification model is established through a machine learning method. In addition, historical vehicle speed data of the third party map service can be utilized, and the collected steering data are combined to judge whether the low-speed steering behavior is caused by congestion or caused by a construction area or an accident.

After the step 43, the embodiment of the present application may further send the type and the position coordinates (such as GPS coordinates) of the first dynamic interest point to the relevant vehicle, so as to prompt the driving safety of the relevant vehicle. Wherein the associated vehicle comprises at least one of: the distance between the first dynamic interest point and the vehicle is smaller than a preset distance, and other member vehicles in a vehicle team where the first vehicle is located; a navigation path passes through the vehicle of the first dynamic point of interest. After the related vehicle receives the information of the type and the position coordinates of the first dynamic interest point, navigation can be planned again or lane changing prompt and the like can be performed in advance, for example, pre-judgment and control can be performed in advance, for example, a heavy truck which is driven automatically can make lane combining decisions in advance by 100 meters.

Fig. 5 further shows a specific example of vehicle-side data acquisition, including:

the data acquisition device continuously monitors the vehicle turn signal and the steering wheel angle. When the turn-on of the steering lamp or the turning of the steering wheel is detected to be larger than a certain preset angle, the steering is judged to be started, and at the moment, relevant data (GPS, steering wheel turning, speed, acceleration and the like) for identifying the dynamic PoI during the steering are collected. When the steering lamp is turned off or the steering wheel is restored to the steering angle smaller than another preset value and kept for a preset period of time, judging that the steering is completed, judging whether the maximum steering angle of the vehicle in the steering process is larger than or equal to 90 degrees, if yes, deleting the related data for identifying the dynamic PoI collected during the current steering, otherwise further judging whether the steering position is on a ramp, if yes, deleting the related data for identifying the dynamic PoI collected during the current steering, if not, storing the related data for identifying the dynamic PoI collected during the current steering at the current vehicle, and sending the data and the vehicle ID to a server through a communication module when the vehicle speed is larger than the typical average speed of the road type. In the above example, when the vehicle passes through a left turn, right turn, or U-turn (steering angle is typically greater than or equal to 90 degrees), or a turn occurs at a high speed highway ramp, the turn is typically less relevant to the dynamic point of interest, and thus the data associated therewith is deleted. In addition, the present example transmits the collected interest point feature data to the server only after the vehicle resumes normal running, so as to avoid occupation of the vehicle computing resources caused by transmitting data in the case of vehicle congestion.

FIG. 6 further illustrates a specific example of identification of server-side dynamic points of interest, including:

the server receives the related data for identifying the dynamic PoI sent by the vehicle (example id=n). The server may incorporate the data of the same vehicle id=n, extract the steering behavior feature data (such as the cumulative number of steering times per unit time, the steering occurrence position, etc.), and then input the steering behavior feature data to the first model (driving behavior judgment model). The first model judges whether the vehicle is a first lane change type (ordinary lane change behavior) or a second lane change type (temporary lane change behavior), excludes the newly added dynamic PoI when judging that the vehicle is a temporary lane change behavior, and inputs relevant data for identifying the dynamic PoI into the second model (PoI judgment model) by using the vehicle id=n at the moment, obtains the PoI type and the position (GPS) output by the second model and sends the obtained data to the relevant vehicle. The associated vehicle may be: vehicles near the point of interest, for example in the range of 200 meters; or other member vehicles of the automatic driving fleet, are traveling to the area near the dynamic point of interest according to the path plan.

Fig. 7 shows an application effect schematic diagram of the method according to the embodiment of the present application, when the method is applied to a scene of identifying interest points in a highway construction area, by the method for collecting feature data of identifying interest points reported by a vehicle according to the embodiment of the present application, feature data collected during a period when the vehicle turns to an adjacent lane is sent to a server, and a first model at the server side determines that "the behavior is not a common lane change behavior"; and judging 'the construction confluence region of the expressway' through a second model, and transmitting the interest point 'the construction confluence region' and GPS coordinates (precision, latitude) to nearby vehicles by the server. By the method, the C vehicle can receive the notification that the road construction exists in the passing area, the position is in the (longitude value and latitude value) state, the interesting point is judged to be 1000m in front of the C vehicle, the C vehicle is a heavy truck, accidents are easy to occur when the C vehicle enters an adjacent lane in a merging mode, and particularly the accident risk of observing a blind area caused by different vehicle types is correspondingly further increased when the nearest vehicle in the adjacent lane is a small-sized passenger car. Therefore, the C vehicles make parallel road behaviors 100m in advance, smoothly pass into the confluent lanes, and simultaneously reduce accident risks.

Based on the method, the embodiment of the application also provides a device for implementing the method.

Referring to fig. 8, an embodiment of the present application further provides a data acquisition device 70, including:

an acquisition module 71 for acquiring and storing steering behavior related data of the vehicle during occurrence of steering behavior;

a judging module 72, configured to judge whether the steering behavior meets a preset condition after the steering behavior is completed;

and a sending module 73, configured to send, to a server, the steering behavior related data during the occurrence of the steering behavior, if the steering behavior meets a preset condition.

Here, the steering behavior-related data includes at least one of the following data:

vehicle track data including track start time, end time, track length, geographic location point timing of the track;

steering behavior data including steering duration, steering angle of steering wheel, geographic location when turning on and off steering lights;

speed-related data including a vehicle speed during occurrence of steering behavior, an opening degree when an accelerator pedal opening degree is greater than a first threshold value, an opening degree when a decelerator pedal opening degree is greater than a second threshold value, a duration for which a speed is lower than a third threshold value;

time and environmental data including date, time of day, weather information and temperature.

Optionally, the preset condition includes at least one of the following conditions: the steering angle is smaller than a preset angle value; the steering position is not a preset area.

As shown in fig. 9, the apparatus further includes:

and a deleting module 74, configured to delete the steering behavior related data during the occurrence of the steering behavior if the steering behavior does not meet a preset condition.

And the sending module is also used for sending the vehicle identification and the steering behavior related data during the occurrence period of the steering behavior to the server when the vehicle speed is greater than the average vehicle speed corresponding to the current road type of the vehicle.

Referring to fig. 10, the embodiment of the present application further provides a server 90, including:

a receiving module 91 for receiving first steering behavior related data during occurrence of steering behavior transmitted by a first vehicle;

the first recognition module 92 is configured to input the first steering behavior related data to a first model for recognizing a lane change behavior type, and obtain a lane change behavior type corresponding to the first steering behavior related data, where the lane change behavior type includes a first lane change type for improving a vehicle speed and a second lane change type for triggering a lane change behavior due to a dynamic interest point;

the second identifying module 93 is configured to input the first steering behavior related data to a second model for classifying the dynamic interest points, and obtain a type and a position coordinate of the first dynamic interest point output by the second model, where the type of the first steering behavior related data corresponds to the second lane change type.

As shown in fig. 11, the server further includes:

and the sending module 94 is configured to send the type and the position coordinates of the first dynamic interest point to the related vehicle. Wherein the associated vehicle comprises at least one of: the distance between the first dynamic interest point and the vehicle is smaller than a preset distance, and other member vehicles in a vehicle team where the first vehicle is located; a navigation path passes through the vehicle of the first dynamic point of interest.

A first training module 95, configured to perform supervised machine learning model training by using pre-collected steering behavior related data corresponding to different lane change types, to obtain the first model;

a second training module 96, configured to perform supervised machine learning model training using pre-collected steering behavior related data corresponding to different dynamic points of interest, to obtain the second model.

It should be noted that, the data acquisition device and the server in this embodiment are devices corresponding to the data acquisition method and the dynamic interest point identification method, and the implementation manners in the foregoing embodiments are all applicable to the embodiments of the devices, so that the same technical effects can be achieved. The device provided by the embodiment of the application can realize all the method steps realized by the embodiment of the method and can achieve the same technical effects, and the parts and the beneficial effects which are the same as those of the embodiment of the method in the embodiment are not described in detail.

The embodiment of the application also provides a dynamic interest point identification system which comprises the data acquisition device and a server.

The embodiment of the application also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements each process of each method embodiment described above, and can achieve the same technical effects, so that repetition is avoided, and no further description is given here. Wherein the computer readable storage medium is selected from Read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic disk or optical disk.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present application.

The embodiments of the present application have been described above with reference to the accompanying drawings, but the present application is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present application and the scope of the claims, which are to be protected by the present application.

Claims (13)

1. A method of data acquisition, comprising:

collecting and storing steering behavior related data of the vehicle during the occurrence of steering behavior;

after the steering behavior is finished, judging whether the steering behavior accords with a preset condition;

and under the condition that the steering behavior accords with a preset condition, transmitting steering behavior related data during the occurrence of the steering behavior to a server.

2. The method of claim 1, wherein the steering behavior related data comprises at least one of:

vehicle track data including track start time, end time, track length, geographic location point timing of the track;

steering behavior data including steering duration, steering angle of steering wheel, geographic location when turning on and off steering lights;

speed-related data including a vehicle speed during occurrence of steering behavior, an opening degree when an accelerator pedal opening degree is greater than a first threshold value, an opening degree when a decelerator pedal opening degree is greater than a second threshold value, a duration for which a speed is lower than a third threshold value;

time and environmental data including date, time of day, weather information and temperature.

3. The method of claim 1, wherein the preset conditions include at least one of: the steering angle is smaller than a preset angle value; the steering position is not a preset area.

4. The method as recited in claim 1, further comprising:

and deleting the steering behavior related data during the occurrence of the steering behavior under the condition that the steering behavior does not accord with a preset condition.

5. The method of claim 1, wherein the sending the steering behavior related data during the occurrence of the steering behavior to a server comprises: and when the vehicle speed is greater than the average vehicle speed corresponding to the current road type of the vehicle, transmitting the vehicle identification and the steering behavior related data during the occurrence of the steering behavior to a server.

6. A method for identifying a dynamic point of interest, comprising:

receiving first steering behavior related data sent by a first vehicle during occurrence of steering behavior;

inputting the first steering behavior related data into a first model for identifying a lane change behavior type, and obtaining a lane change behavior type corresponding to the first steering behavior related data, wherein the lane change behavior type comprises a first lane change type for improving the vehicle speed and a second lane change type for triggering the lane change behavior due to the existence of a dynamic interest point;

and under the condition that the corresponding track behavior type of the first steering behavior related data is the second track changing type, inputting the first steering behavior related data into a second model for classifying the dynamic interest points, and obtaining the type and the position coordinates of the first dynamic interest points output by the second model.

7. The method as recited in claim 6, further comprising:

transmitting the type and the position coordinates of the first dynamic interest point to a related vehicle, wherein the related vehicle comprises at least one of the following: the distance between the first dynamic interest point and the vehicle is smaller than a preset distance, and other member vehicles in a vehicle team where the first vehicle is located; a navigation path passes through the vehicle of the first dynamic point of interest.

8. The method as recited in claim 6, further comprising:

performing supervised machine learning model training by utilizing pre-collected steering behavior related data corresponding to different lane change types to obtain the first model;

and performing supervised machine learning model training by utilizing steering behavior related data which are collected in advance and correspond to different dynamic interest points, so as to obtain the second model.

9. A data acquisition device, comprising:

the acquisition module is used for acquiring and storing steering behavior related data of the vehicle during the occurrence of steering behavior;

the judging module is used for judging whether the steering behavior accords with a preset condition after the steering behavior is finished;

and the sending module is used for sending the steering behavior related data during the occurrence period of the steering behavior to a server under the condition that the steering behavior accords with a preset condition.

10. A server, comprising:

the receiving module is used for receiving first steering behavior related data sent by the first vehicle during the occurrence of steering behavior;

the first recognition module is used for inputting the first steering behavior related data into a first model for recognizing the lane change behavior type, and obtaining the lane change behavior type corresponding to the first steering behavior related data, wherein the lane change behavior type comprises a first lane change type for improving the vehicle speed and a second lane change type for triggering the lane change behavior due to the existence of a dynamic interest point;

the second recognition module is used for inputting the first steering behavior related data into a second model for classifying the dynamic interest points under the condition that the behavior type corresponding to the first steering behavior related data is a second lane change type, and obtaining the type and the position coordinates of the first dynamic interest points output by the second model.

11. The server according to claim 10, further comprising:

the sending module is used for sending the type and the position coordinates of the first dynamic interest point to a related vehicle, wherein the related vehicle comprises at least one of the following components: the distance between the first dynamic interest point and the vehicle is smaller than a preset distance, and other member vehicles in a vehicle team where the first vehicle is located; a navigation path passes through the vehicle of the first dynamic point of interest.

12. The server according to claim 10, further comprising:

the first training module is used for training a supervised machine learning model by utilizing steering behavior related data which are collected in advance and correspond to different lane change types to obtain the first model;

and the second training module is used for performing supervised machine learning model training by utilizing the steering behavior related data which are collected in advance and correspond to different dynamic interest points, so as to obtain the second model.

13. A dynamic point of interest identification system comprising at least one data acquisition device according to claim 9, and a server according to any one of claims 10 to 12.